Overreaction and Investment Choices: An Experimental Analysis

We study the degree of overreaction and the relation of overreaction and psychological biases as well as financial consequences of overreaction in a controlled experimental setting with 104 participants. The majority of participants tend to overreact, however, the degree of overreaction is heterogeneous. A few subjects even underreact. We also measure the overconfidence of the participants with a miscalibration scale. In line with theoretical predictions we find that more overconfident subjects overreact more. We also find that overreaction is associated with higher levels of risk taking after good signals and lower levels of risk taking after bad signals. Finally, overreaction harms portfolio efficiency, as measured by the Sharpe ratio.